Computer-Aided Drug Design: from Discovery of Novel Pharmaceutical Agents to Systems Pharmacology
Vladimir Poroikov
Abstract
New drug discovery is based on analysis of existing information about the pathological mechanisms, molecular targets, and ligands, which can interact with particular targets thus causing a pharmacotherapeutic effect. The available data on diseases, drugs, pharmacological effects, molecular targets, and drug-like substances, taking into account the combinatorics of the associative relationships between them, correspond to the Big Data. Analysis of such data requires application of modern computer-aided drug design methods. This review summarizes results obtained by the Laboratory for Structure-Function Based Drug Design, Institute of Biomedical Chemistry, in this area. We have developed approaches for identification of promising pharmacological targets, prediction of several thousand types of biological activity based on the structural formula of the compound, analysis of protein-ligand interactions considering local similarity of amino acid sequences, identifying of putative molecular mechanisms responsible for side effects of drugs, calculation of the integral toxicity of drugs taking into account their metabolism, in the human body, as well as prediction of resistant and sensitive strains and evaluation of the effectiveness of therapy by combinations of antiretroviral drugs in view of the molecular genetic peculiarities of the clinical isolates of HIV-1. Our computer programs are implemented as the web-services freely available in the Internet; they are used by thousands of researchers from many countries of the world to select the most promising substances for synthesis and to identify the priority areas for experimental testing of their biological activity.